Table 2 Postintervention means (95% CI) and noninferiority results for primary, secondary and exploratory outcomes at 3 months post-randomization

From: Task-sharing and telemedicine delivery of psychotherapy to treat perinatal depression: a pragmatic, noninferiority randomized trial

Outcome (measure; range)

Nonadjusted mean¥ (95% CI)

95% CI

upper bound±

NIM*

Adjusted mean¥ (95% CI)

95% CI

upper bound±

NIM*

Primary and secondary outcomes

 ITT: nonspecialist versus specialist (10% NIM)

Nonspecialist provider

Specialist provider

  

Nonspecialist provider

Specialist provider

  

 Depressive symptoms (EPDS, 0–30)

9.27 (8.85–9.70)

8.91 (8.49–9.33)

0.86

0.89*

8.66 (7.85–9.46)

8.33 (7.51–9.14)

0.77

0.89*

 Anxiety symptoms (GAD-7, 0–21)

6.44 (6.01–6.86)

6.36 (5.95–6.78)

0.57

0.64*

6.04 (5.23–6.84)

6.17 (5.36–6.98)

0.31

0.64*

 ITT: telemedicine versus in-person (13% NIM)

Telemedicine

In-person

  

Telemedicine

In-person

  

 Depressive symptoms (EPDS, 0–30)

9.15 (8.79–9.50)

8.92 (8.38–9.45)

0.77

1.16*

8.61 (7.83–9.39)

8.37 (7.50–9.23)

0.78

1.16*

 Anxiety symptoms (GAD-7, 0–21)

6.43 (6.09–6.78)

6.29 (5.71–6.88)

0.73

0.82*

6.14 (5.34–6.93)

6.02 (5.15–6.90)

0.65

0.82*

 PP: telemedicine versus in-person (13% NIM)†

Telemedicine

In-person

  

Telemedicine

In-person

  

 Depressive symptoms (EPDS, 0–30)

9.17 (8.82–9.53)

8.81 (8.26–9.36)

0.91

1.15*

8.56 (7.77–9.34)

8.22 (7.34–9.10)

0.89

1.15*

 Anxiety symptoms (GAD-7, 0–21)

6.47 (6.12–6.81)

6.17 (5.58–6.76)

0.90

0.80

6.20 (5.42–6.98)

5.86 (4.98–6.74)

0.89

0.80

Exploratory outcomes

 ITT: nonspecialist versus specialist (10% NIM)

Nonspecialist provider

Specialist provider

  

Nonspecialist provider

Specialist provider

  

 Quality of life (EQ-5D-5L domains, 1–5)

 Mobility

1.35 (1.29–1.41)

1.33 (1.27–1.38)

0.09

0.13*

1.34 (1.22–1.45)

1.31 (1.20–1.43)

0.08

0.13*

 Self-care

1.24 (1.19–1.28)

1.25 (1.20–1.29)

0.04

0.13*

1.19 (1.10–1.28)

1.22 (1.12–1.31)

0.02

0.13*

 Usual activities

1.74 (1.67–1.81)

1.71 (1.63–1.77)

0.12

0.17*

1.69 (1.55–1.84)

1.64 (1.50–1.78)

0.13

0.17*

 Pain/discomfort

1.91 (1.84–1.99)

1.87 (1.80–1.94)

0.13

0.19*

1.87 (1.73–2.02)

1.83 (1.68–1.97)

0.12

0.19*

 Anxiety/depression

2.27 (2.19–2.34)

2.26 (2.18–2.33)

0.09

0.23*

2.06 (1.92–2.21)

2.08 (1.94–2.23)

0.06

0.23*

 Disability (WHODAS 2.0, 12–60)

20.93 (20.24–21.63)

20.79 (20.14–21.45)

0.94

2.08*

20.62 (19.47–21.77)

20.47 (19.31–21.63)

0.78

2.08*

 ITT: telemedicine versus in-person (13% NIM)

Telemedicine

In-person

  

Telemedicine

In-person

  

 Quality of life (EQ-5D-5L domains, 1–5)

 Mobility

1.34 (1.29–1.39)

1.33 (1.26–1.41)

0.08

0.17*

1.31 (1.20–1.42)

1.34 (1.22–1.46)

0.04

0.17*

 Self-care

1.24 (1.20–1.28)

1.24 (1.18–1.31)

0.06

0.16*

1.20 (1.11–1.29)

1.21 (1.11–1.31)

0.05

0.16*

 Usual activities

1.71 (1.65–1.77)

1.76 (1.66–1.86)

0.05

0.23*

1.63 (1.49–1.76)

1.71 (1.55–1.86)

0.01

0.23*

 Pain/discomfort

1.89 (1.83–1.95)

1.91 (1.81–2.01)

0.08

0.25*

1.83 (1.70–1.97)

1.87 (1.71–2.02)

0.06

0.25*

 Anxiety/depression

2.27 (2.21–2.33)

2.23 (2.13–2.34)

0.15

0.29*

2.09 (1.95–2.23)

2.05 (1.90–2.21)

0.13

0.29*

 Disability (WHODAS 2.0, 12–60)

20.77 (20.22–21.32)

21.16 (20.21–22.12)

−0.39

2.75*

19.94 (18.71–21.17)

20.41 (19.05–21.78)

0.36

2.75*

Outcome (measure; range)

Nonadjusted mean¥ (95% CI)

95% CI lower bound±

NIM*

Adjusted mean¥ (95% CI)

95% CI lower bound±

NIM*

 ITT: nonspecialist versus specialist (10% NIM)

Nonspecialist provider

Specialist provider

  

Nonspecialist provider

Specialist provider

  

 Client satisfaction (CSQ-8, 1–4)

3.40 (3.34–3.46)

3.44 (3.39–3.49)

−0.10

−0.34*

3.43 (3.32–3.55)

3.51 (3.40–3.62)

−0.14

−0.34*

 Therapeutic alliance (WAI–SR, 1–5)

4.20 (4.13–4.27)

4.12 (4.05–4.19)

−0.01

−0.41*

4.08 (3.93–4.23)

4.07 (3.92–4.22)

−0.07

−0.41*

 Patient activation (PAAS, 0–4)

2.17 (2.09–2.24)

2.13 (2.06–2.21)

−0.06

−0.21*

2.11 (1.96–2.26)

2.07 (1.92–2.22)

−0.04

−0.21*

 Perceived support (MSPSS, 1–7)

5.53 (5.43–5.62)

5.46 (5.35–5.57)

−0.05

−0.55*

5.50 (5.32–5.67)

5.43 (5.26–5.61)

−0.03

−0.55*

 ITT: telemedicine versus in-person (13% NIM)

Telemedicine

In-person

  

Telemedicine

In-person

  

 Client satisfaction (CSQ-8, 1–4)

3.43 (3.39–3.47)

3.38 (3.29–3.46)

−0.13

−0.44*

3.53 (3.42–3.64)

3.41 (3.30–3.53)

0.04

−0.44*

 Therapeutic alliance (WAI–SR, 1–5)

4.17 (4.11–4.22)

4.13 (4.02–4.24)

−0.14

−0.54*

4.13 (3.98–4.28)

4.02 (3.86–4.18)

0.01

−0.54*

 Patient activation (PAAS, 0–4)

2.15 (2.09–2.21)

2.17 (2.06–2.27)

−0.08

−0.28*

2.07 (1.93–2.21)

2.11 (1.95–2.27)

−0.14

−0.28*

 Perceived support (MSPSS, 1–7)

5.48 (5.40–5.56)

5.54 (5.39–5.69)

−0.08

−0.72*

5.45 (5.28–5.62)

5.48 (5.29–5.67)

−0.15

−0.72*

  1. *Noninferiority was met.
  2. ¥Per the SAP, nonadjusted means represent nonimputed analyses without covariates. Outliers for depressive and anxiety symptoms can be found in Supplementary Table 8. Imputed analyses can be found in Supplementary Table 9. Regression models include imputed data and covariates. Covariates: perinatal period, hub, age, baseline EPDS and GAD-7, race/ethnicity, employment, modality and provider preference, history of seeing a therapist, history of depression/anxiety, years lived in current home, nulliparous, time between randomization and 3 months, psychotropic medication before 3 months, started psychotropic medication after enrollment, pregnancy conditions, modality and provider (Supplementary Table 7). CSQ-8, Client Satisfaction Questionnaire60; EQ-5D-5L, EQ-5D 5-level61; MSPSS, Multidimensional Scale of Perceived Social Support62; NIM, noninferiority margin; NSP, nonspecialist provider; PAAS, Premium Abbreviated Activation Scale63; SP, specialist provider; WAI–SR, Working Alliance Inventory–Short Revision64; WHODAS, World Health Organization Disability Assessment Schedule 2.0 (ref. 65).
  3. †Per protocol analyses refer to the condition (telemedicine versus in-person) that the participant actually received and only apply to modality because no participants were switched between provider conditions. Of the 21 protocol deviations, 16 received telemedicine and 5 received in-person and were analyzed in the PP analyses accordingly.
  4. ±The 95% upper bound is used when higher scores are worse and the 95% lower bound is used when lower scores are worse.
Back to article page